A rag + rag-graph native code intelligence layer for Java microservice estates, exposed to LLM agents via CLI
Latest release v0.12.2 · 13 Sept 2026
16 files for Claude Code, Codex and OpenCode: docs-watcher, brainstorming, dispatching-parallel-agents, executing-plans and 12 more — 950 tokens loaded in every session.
AGENTS.md C 950 tok .agents/skills/brainstorming/SKILL.md A 38 tok .agents/skills/dispatching-parallel-agents/SKILL.md A 26 tok .agents/skills/executing-plans/SKILL.md A 22 tok .agents/skills/finishing-a-development-branch/SKILL.md A 44 tok .agents/skills/publish-pip/SKILL.md C 73 tok .agents/skills/receiving-code-review/SKILL.md A 38 tok .agents/skills/requesting-code-review/SKILL.md A 22 tok .agents/skills/subagent-driven-development/SKILL.md A 17 tok .agents/skills/systematic-debugging/SKILL.md A 21 tok .agents/skills/test-driven-development/SKILL.md A 17 tok .agents/skills/using-git-worktrees/SKILL.md A 36 tok .agents/skills/using-superpowers/SKILL.md A 28 tok .agents/skills/verification-before-completion/SKILL.md A 43 tok .agents/skills/writing-plans/SKILL.md A 21 tok .agents/agents/docs-watcher.md A 86 tok These files are HumanBean17/jrag's own configuration — they tell Claude Code, Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.